Přejít k hlavnímu obsahu


Conference Paper (international conference)

Exploitation of particle filter for forecast error covariance structure estimation

Hofman Radek

: Proceedings of the 9th International PhD Workshop on Systems and Control, Young Generation Viewpoint, p. 1-6 , Eds: Gašperin Matej, Pregelj Boštjan

: 9th International PhD Workshop on Systems and Control: A Young Generation Viewpoint, (Izola, SI, 01.10.2008-03.10.2008)

: CEZ:AV0Z10750506

: GA102/07/1596, GA ČR

: data assimilation, estimation, marginalized, particle filter, model error covariane structure

: http://library.utia.cas.cz/separaty/2008/AS/hofman-exploitation%20of%20particle%20filter%20for%20forecast%20error%20covariance%20structure%20estimation.pdf

(eng): The paper presents a scheme for estimation of spatio-temporal evolution of a quantity with unknown model error based on LU model. Solution of LU model is efficient and easy to implementation even for high dimensional problems. Model error is assumed to be uniformly distributed and it is estimated upon measured and modeled values. Methods of linear programming are applied to the problem. The main contribution of this paper is application of general LU estimation algorithm to the linear--uniform problem with unknown model error magnitude. Methodology is demonstrated on the problem of modeling of spatio--temporal evolution of groundshine--dose from radionuclides deposited on terrain in long--time horizon. Achieved results and the methodology is compared to the results obtained by an approach based on the MPF algorithm. The advantages of particular methods are concluded.

(cze): článek se zabývá schématem pro odhadování spatio-temporální evoluce kvantity s neznámým modelem chyb založeném na modelu LU.

: BC